Editing ATF with Nammu

Introducing Nammu, Oracc's new ATF editor

Nammu is Oracc's new, platform-independent editing software for creating transliterations, lemmatisations and translations of cuneiform texts. It is being written for us by Raquel Alegre and Jens Nielsen of UCL's Research Software Development Team.

• Download the current release of Nammu
• Follow these installation instructions

If you are an existing Oracc content-creator and have been editing ATF in Emacs, that's all you need to do. You'll find that Nammu is an intuitive replacement for Emacs, and that you can use all the Oracc keyboard shortcuts as usual.

New to ATF?

If you haven't edited ATF before, you may also want to download:

• our keyboard for cuneiform transliteration
• and/or one or more unicode transliteration fonts

You should also read the online help pages on learning and using Oracc ATF and on lemmatising.

Do contact any one of the Oracc Steering Group (Eleanor, Jamie, Niek, or Steve) if you would like to have your own Oracc project space (which can be public or private, as you like).

Tips and tricks

Nammu now has all its core funcationality – you can edit real texts with it unproblematically -- but it is still under development. If you like you can join the oracc/nammu group on Github to:

• watch for new releases
• see which aspects of Nammu are currently being worked on, and
• most helpfully for us, report bugs and issues.

Here are some ease-of-use features that we're currently working to enhance:

• At the moment there is no line-wrapping, so you have to scroll across for long lines.
• Nor is there any search/replace facility yet — also coming soon, we hope.
• There's no split screen option yet, either, so for the moment I suggest that you create translations in your usual word processor as you lemmatise, then paste them into Nammu for validation when you finish. It works for me!

For more details see here.

More about Nammu

Nammu has some built-in help but you can also:

• watch a video of me using Nammu to edit a tablet from Tell Khaiber
• watch the whole video, in which I introduce Oracc and Tell Khaiber
• read Raquel's slides about the programming and principles behind Nammu
• ask any one of the Oracc Steering Group (Eleanor, Jamie, Niek, or Steve) for further help!

Adding a Twitter feed to your ESP portal

for project managers

If you have a Twitter account, and your Oracc project has an ESP portal site, you can now add a Twitter timeline feed to it by following the instructions here.

You can see an example of it in use in the new Nimrud project portal (still tiny but growing fast).

In other Oracc news:

• The search functionality, which had been broken for a few days, is now up and running again.
• The GKAB portal pages have now moved closer to the CAMS/GKAB corpus; please update your bookmarks.

Installing EmacsW32 on Windows 7

The Emacs installation documentation at http://oracc.museum.upenn.edu/doc/builder/emacs/ now includes instructions for installing EmacsW32 on Windows 7 (shown in bold). Do let us know if there are any leftover problems with it.

Updated documentation

Over the past week or so I've had a jolly time checking that all the Oracc documentation is up-to-date and reflects current practice, and that all the links work. I won't list every tiny change here, but would like to draw your attention to a few points in particular:

All users: Citing Oracc URLs online and in print

Oracc 2 makes systematic use of stable URIs, so that you can be confident that print references and online links will remain valid. That means, though, that some details of the instructions on how to cite Oracc pages and resources have changed.

Mostly for editors and managers: Oracc language codes

If you have already read my overview of how the new lemmatiser works, you'll know that now you always need to specify Akkadian dialects (unless you are producing ATF for CDLI) and that glossary merges must now always specify the language. I've also worked those changes systematically through the documentation.

For editors and project managers: builder documentation

Editors and project managers will also notice that I have rearranged the list of builder documentation. It's still all there as before but now, I hope, in a more logical order.

For project managers: glossary management

I have incorporated the L2 updates into the sections on glossaries in the pages on project management with Unix and project management with Emacs (and written it into the rest of the documentation).

For project managers: The Oracc Command

This page has been reorganised and updated, so that it is easier to find help on the commands you most need for everyday project management work.

And you'll now find a Documentation full text search at the top of every documentation page.

If you notice anything wrong, broken, missing, or just plain confusing, please let me know, either in the comments or by email. The documentation is for you, the users and project creators, so please tell me if (and how) it isn't serving your needs.

ATF on Emacs

(For editors and project managers)

We still need to do a little work on updating Emacs ATF mode for Oracc 2. Until that happens, here is a list of what is and isn't working, and how to work around the latter.

ATF menu

Available when you have a .atf file open.

Check ATF

Lemmatise Buffer

Create Template

ATF Mode help

All working as usual.

Harvest Notices

Count Lemmata

Not working at the moment but neither is essential to the editing process.

Browse ATF docs

Links to a dead page. Go to the Builder Pages instead.

CBD menu

Corpus-based dictionary menu for project managers; vailable when you have a .glo file open.

Check CBD

Lemmer Update

Rebuild

Not working at the moment; you should manage glossaries through Unix instead

Project Maintenance Mode

CBD Mode help

Working as usual

Browse CBD docs

Links to a dead page. Go to the Manager Pages instead.

ORACC menu

For project managers; accessible from the Project Maintenance Mode item on the CBD menu.

Check Project

Merge Glossaries

Not working at the moment; you should manage projects through Unix instead—and note that the merge command now needs to specify the language

Harvest Glossary

Working as usual

Rebuild Project

Working but no longer recommended; use ncr or nor in Unix instead

Please do let me know if you have any problems or questions.

Adding Full Text Search to your Oracc 2 project portal

(FOR ORACC PROJECT MANAGERS)

One of Oracc 2's new functionalities is the Full Text Search across the whole of the site. There is a new documentation page about how to use it.

It is also very easy to enable users to do Full Text Search on your project alone, from one or more portal pages of your project. To do so you will need to edit them with Emacs or an XHTML editor.

• First add this line of code somewhere within the head element of the page, ideally with the other link elements:

  <link rel="stylesheet" type="/css/oraccbase.css"/>

  EDIT: This is now built into the main Oracc stylesheet.

• Then add this code wherever you want it to appear in the body of your web page, replacing [PROJECT] with the name of your project:

  <span class="find-button"><a href="/[PROJECT]/find">Full 
  Text Search</a></span>

  For instance, you might add a link to the online help:

  <p>Do a <span class="find-button"><a href="/cams/find">Full 
  Text Search</a></span> in CAMS 
  <a href="/doc/user/fulltextsearch/">[HELP]</a>.</p>

  EDITED TO ADD:

  If you are working with ESP, use <esp:link url="/[PROJECT]/find">...</esp:link> instead of <a href="/[PROJECT]/find">...</a>. The necessary CSS code is already built into ESP's stylesheet.

  This is displayed in the browser as:

  Do a Full Text Search in CAMS [HELP].

• Finally, don't forget to run oracc web to make it go live online.

Note that this does not replace the existing text, catalogue and glossary search functions, but allows users to search across the whole of your project, including portal pages. That does mean though that it is not very sensitive to the type of data it is searching, and returns rather rough and ready results. In particular, at the moment it privileges glossary entries but Steve is improving the way it returns results all the time.

As always, please let me know if you need assistance.

New documentation

As the new-look Oracc comes onstream, you will find that it looks a little bit different, and that you can do a lot more with the texts than before. In particular, Steve has dramatically improved the search functionality. People making projects and corpora on Oracc will also, we hope, notice several significant improvements.

I have updated the online help pages for Oracc website users and for project creators. Please take the time to read the pages are most relevant to you, as they will help you get a great deal more out of Oracc. If you have any questions or problems, or can't find what you're looking for, please leave a comment here for me.

Links to help from the Oracc homepage will be reinstated shortly. What you see at the moment is just temporary, while the upgrade is in progress. Meanwhile they are here on the right.

For all Oracc website users

• Using the new-look Oracc 2 online corpora

You can also access this page from the Help button at the top of all Oracc corpus pages.

Note that this is a different URL to the old help pages, which I haven't yet taken down. So if you have bookmarked the old ones, please update your bookmarks. The main user help page has been updated though.

For Oracc project creators

• L2: how to use it

A one-page starting point for all editors and project managers, with links to all the relevant new documentation. You may not need all of this if you don't run a whole project single-handedly; just stop reading when it no longer seems relevant.

Not much has changed in fact, especially for text editing. You will find that lemmatisation is much improved; that you have much more control over individual languages, dialects and glossaries; and that projects rebuild much faster than they did before.

If your project has not yet been migrated it will be soon. We will email you when we are ready for your input. As the new system is much more rigorous than the old one, you will almost certainly have to spend some time error-checking and bug-fixing before you can continue routine work. Be prepared! Remember too that you will need to think of a new password.

Documentation updates

We've added a few useful odds and ends to the Oracc documentation over the past few months, which I thought it would be useful to summarise here:

• How to report a new sign value to Niek when the ATF checker complains of an unknown value;
• How to search the Qcat (catalogue of compositions) and request new entries;
• A list of Akkadian roots;
• How to add linguistic annotations to project glossaries;
• Fuller information about discourse divisions;
• And several new options for the oracc command.

As always, just ask if you have any questions or problems - or any suggestions for correcting, improving or adding to the documentation.

NEW: portal site software

Would you like to build a portal website for your Oracc project along the lines of Knowledge and Power or Assyrian Empire Builders? Now you can!

Just this week we have finished testing and documenting ESP, an Oracc version of the electrostatic programme written by George McKerron for the Whipple Museum of the History of Science in Cambridge. In order to use it, you will need to know the basics of (X)HTML and how to use Emacs (but most of us use Emacs for Oracc text editing anyway).

Instructions for use are on the Oracc website. The range of options for Oracc portal sites (including but not only ESP) is here.

As always, if you would like any advice or help, just ask me.

Qcat overhauled

The Qcat—the catalogue of compositions—has now been radically rationalised, tidied up and documented. In particular, the naming conventions for the designations (the prose equivalents of the Q-numbers) have been standardised.

If your project uses the Qcat, you may find that you will need to revise your project's Q designations accordingly. If you use a Filemaker database or similar for your catalogue, I can send you a list of new designations to import, overwriting the old ones. Just email me.

If and when your project needs new Q-numbers, send me a list of your proposed designations and other relevant fields (as per the documentation); or ask me for a clone of the Filemaker Qcat database, to fill in and email back to me.

If you ever notice mistakes, duplicates or omissions in the Qcat, or have any questions about it, please email me about that too.

A note about configuration options

Looking over the posts on the blog and a few e-mails I realized that some comments about the configuration file might be helpful.

The configuration system is in flux and has grown rapidly and organically, particularly over the last nine months. Oracc software can access the configuration file, config.xml, via several interfaces, and some of the programs ignore options that don´t apply to them while others complain about misunderstood or illegal options.

In general, if you see an error message about a configuration option, just check that you haven´t mistyped an option or value name. Otherwise, leave suspicious options in the config.xml because some program or other may be secretly using them.

If you change 00lib/config.xml you need to use at least:

oracc config

To update the installed version. You may need to do:

oracc web

Or even:

oracc rebuild

Before you see the effects of your changes.

After the remaining to-do list items are dealt with, the configuration system will get a complete review. Some things that cannot be configured in config.xml will be added (like the buttons SAAo uses); others will be documented; others will be rationalized as there are a few places where multiple options do very similar things.

I am also hoping (expecting, even) to create a web interface with an easy form for managing config.xml that has the documentation built in to it and drop-downs with the valid option values etc.

So, this pain with config.xml is transient. I am expecting to have the new configuration system in place by the end of September.

Directory names have changed!

If you are struggling to do project management through emacs, it may be because you are still using the old (CDL) directory names. For instance:

• old sources/ is now 00atf/
• old lib/ is now 00lib/
• old cat/ is now 00cat/
• old websources/ is now 00web/
• old backups/ is now 00bak/
  

See Project Management through Emacs for step-by-step instructions and Project Files for a more systematic overview.

As always, do ask for help if you need it, but I may not be as speedily responsive as usual over the next few days (up to and including Wednesday), as I'll be doing tablet work in Berlin. I will have broadband access in the hotel though, so I'll be checking in at least twice a day.

New doco structure

As you'll see, the Oracc documentation is much fuller and more systematically organised than the old CDL doco. It's fully searchable from the Oracc home page.

You'll also find that visitors to your website can access all the user help from the [HELP] link at the left of every Page View and Item View header on your corpus pages.

But if you think anything is missing, or needs further explanation, please let me know.